The prediction of the critical factor of safety of homogeneous finite slopes using neural networks and multiple regressions

Erzin, Yusuf; Çetin, Tülin

doi:10.1016/j.cageo.2012.09.003

Cited by 122 publications

(38 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The values of root mean square error (RMSE) and variance account for (VAF) are determined for the training and testing datasets. For a good model, the values of RMSE and VAF should be 0 and 100%, respectively (Erzin and Cetin, 2013). For training dataset, the …”

Section: Resultsmentioning

confidence: 99%

Pullout capacity of small ground anchor: a least square support vector machine approach

Samui

Kim

Aiyer

2015

J. Zhejiang Univ. Sci. A

View full text Add to dashboard Cite

This study employs the least square support vector machine (LSSVM) for the prediction of pullout capacity of small ground anchor. LSSVM is firmly based on the theory of statistical learning and uses regression technique. In LSSVM, Vapnik and Lerner (1963)'s ε-insensitive loss function was replaced by a cost function which corresponded to a form of ridge regression. The input parameters of LSSVM were equivalent anchor diameter, anchor embedment depth, average cone tip resistance, average cone sleeve friction, and installation technique. Using 83 out the available 119 in-situ test datasets, an LSSVM regression model was developed. The goodness of the model was tested using the remaining 36 data points. The developed LSSVM also gave an error bar of predicted data. A sensitivity analysis was conducted to determine the effect of each input parameter. The results were compared with the artificial neural network (ANN) model. Overall, LSSVM was shown to perform well.

show abstract

Section: Resultsmentioning

confidence: 99%

Pullout capacity of small ground anchor: a least square support vector machine approach

Samui

Kim

Aiyer

2015

J. Zhejiang Univ. Sci. A

View full text Add to dashboard Cite

show abstract

“…Learning rates of 0.05 and 0.5 were suggested in the studies conducted by Jahed Armaghani et al [42] and Choobbasti et al [79], respectively. Yilmaz and Yuksek [56], Erzin and Cetin [80] and Momeni et al [81] recommended the value of 0.01 for learning rate, while this value was suggested as 0.1 in the study conducted by Yagiz et al [65]. Apart from learning rate, a steadying effect can be observed by momentum coefficient [82].…”

Section: Non-linear Multiple Regression Modelmentioning

confidence: 97%

Application of several non-linear prediction tools for estimating uniaxial compressive strength of granitic rocks and comparison of their performances

Armaghani

Mohamad

Hajihassani

et al. 2015

Engineering with Computers

115

View full text Add to dashboard Cite

tools, namely non-linear multiple regression (NLMR), artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS). After conducting the mentioned models, considering several performance indices including coefficient of determination (R 2 ), variance account for and root mean squared error and also using simple ranking procedure, the models were examined and the best prediction model was selected. It is concluded that the R 2 equal to 0.951 for testing dataset suggests the superiority of the ANFIS model, while these values are 0.651 and 0.886 for NLMR and ANN techniques, respectively. The results pointed out that the ANFIS model can be used for predicting UCS of rocks with higher capacity in comparison with others. However, the developed model may be useful at a preliminary stage of design; it should be used with caution and only for the specified rock types.

show abstract

“…The general purpose of this method is to discover the relationship between several independent or predictor variables and a dependent or criterion variable [16]. In its natural in-situ stress state, the rock mass can be assumed to be stable and its plastic deformation can be assumed to be non-increasing.…”

Section: Methods Fundamental Methodsmentioning

confidence: 99%

A modified initial in-situ Stress Inversion Method based onFLAC3D with an engineering application

Guo

Zhu

et al. 2015

Open Geosciences

View full text Add to dashboard Cite

Abstract:To improve the accuracy of an initial in-situ stress eld determined by inversion, we describe a modi ed initial in-situ stress inversion method that uses partial least-squares regression based on FLAC3D. First, each stress component is regressed to improve the tting accuracy of locally abnormal stress regions, and then the relationship between element stress and unbalanced node force is analyzed according to the computational principles of FLAC3D. The initial in-situ stresses obtained from these regression calculations are added to a numerical model, and the unbalanced node forces are recalculated. An external force equal to the recalculated unbalanced node force is then exerted on the node in the direction opposing the original unbalanced node force to satisfy the equilibrium condition. For the in-situ stresses of elements that do not satisfy the strength conditions, they are modi ed by assuming the average stress is constant and reducing the partial stress to satisfy the equilibrium and strength conditions, which also resolves the unreasonable distribution of the boundary nodal forces and results in good regression estimates. A three-dimensional hypersurface spline interpolation method is developed to calculate the in-situ stress tensor at arbitrary coordinates. Finally, we apply this method to an underground engineering project, and the results are shown to agree well with those obtained from eld monitoring. Therefore, it is concluded that this modi ed in-situ stress inversion method could e ectively improve the tting accuracy of locally abnormal stress regions.

show abstract

The prediction of the critical factor of safety of homogeneous finite slopes using neural networks and multiple regressions

Cited by 122 publications

References 45 publications

Pullout capacity of small ground anchor: a least square support vector machine approach

Pullout capacity of small ground anchor: a least square support vector machine approach

Application of several non-linear prediction tools for estimating uniaxial compressive strength of granitic rocks and comparison of their performances

A modified initial in-situ Stress Inversion Method based onFLAC3D with an engineering application

Contact Info

Product

Resources

About